Initiating a Customer Data Governance Model

While you can initiate data governance without having an MDM practice, you can't have an effective MDM practice without data governance. Data governance is the glue in MDM. Without it, MDM will not be a cohesive sustainable practice. In a 2010 study conducted by The Information Difference,¹ 88 percent of the respondents felt that implementing data governance either prior to or together with MDM and data quality is critical to the success of an MDM initiative.

Defining and planning a Customer Data Governance (CDG) model should start with an initial assessment and the groundwork necessary to drive a sound charter and implementation proposal for your governance model. The assessment and proposal should predominantly be a business-driven initiative with sponsorship at the vice president (VP) level to ensure strong commitment and advocacy exists for establishing data governance. Within the sponsoring organization, there should be an existing director or senior manager appointed by the VP to lead the tactical aspects of the initiative. The sponsoring VP along with this tactical leader will be the driving force needed to get the data governance ball rolling.

Involvement of a consultant in the planning phase of your governance initiative can also be a major benefit. A consultant with a good track record in the MDM and data governance space can greatly assist in best practice evaluations, initial process or data quality assessments, gathering of metadata, evaluation of third-party MDM tools or services, preparation of reference material, or facilitation and planning of meetings. But always keep in mind that CDG needs to be developed as an organic business-driven function within the company. Consultants should only be leveraged in very specific roles to help establish the foundation for CDG or to help accelerate the implementation of CDG. Overuse of consultants will actually become a limiter and an ongoing cost factor that can potentially kill the program if budget constraints or reductions occur.

As a general rule, planning and initiating a CDG model is something that should be done early on, but data governance should never be implemented hastily. A bad model and charter will probably fail. There is an excellent white paper authored by Jill Dyché and Kimberly Nevala of Baseline Consulting entitled “Ten Mistakes to Avoid when Launching Your Data Governance Program.”² We highly recommend you refer to this as you plan your governance initiative. The guidance and practical approaches we cover in this chapter should help to greatly avoid those mistakes. If your governance plan considers and implements a well-balanced approach across the aspects of data ownership, decision making, data stewardship, and data quality and control, a successful governance model will ensue and will be well equipped to handle the operating challenges and course corrections that can emerge in any company.

In the long run, data governance becomes less about the big decisions and macro-level control of the data, and becomes more about institutionalizing the consistent micromanagement of the data. Of course, the decision making, big rules, and the overarching influence of a data governance council are still critical, but over time the influence and mechanics of data governance should settle into the business fabric to form more of a day-to-day operation and quality management process that is comfortably interacting with the various LOB and IT functions.

Let's first take a look at an example of a high-level CDG process design and implementation approach shown in Figure 4.1. This will help explain the overall planning and execution needed to achieve the end state. Then we'll break this down further to discuss the planning and implementation components.

Figure 4.1 Example of CDG Process Design and Implementation Approach